This paper introduces a nested embedding learning framework for Arabic NLP, utilizing Matryoshka Embedding Learning and multilingual models. The authors translated sentence similarity datasets into Arabic to enable comprehensive evaluation. Experiments on the Arabic Natural Language Inference dataset show Matryoshka embedding models outperform traditional models by 20-25% in capturing Arabic semantic nuances. Why it matters: This work advances Arabic NLP by providing a new method and evaluation benchmark for semantic similarity, which is crucial for tasks like information retrieval and text understanding.
Technology Innovation Institute (TII) has appointed Dr. Hakim Hacid as Chief Researcher of its AI and Digital Science Research Center (AIDRC), the home of the Falcon LLM series. Dr. Hacid previously served as Executive Director and Acting Chief Researcher at TII, after joining in 2022 from Zayed University. He brings expertise in AI, ML, data science, and information retrieval, with prior experience at Bell Labs and Macquarie University. Why it matters: The appointment strengthens TII's leadership in AI research and development, particularly for the Falcon series of open-source LLMs that have gained global recognition.
This paper introduces an enhanced Dense Passage Retrieval (DPR) framework tailored for Arabic text retrieval. The core innovation is an Attentive Relevance Scoring (ARS) mechanism that improves semantic relevance modeling between questions and passages, replacing standard interaction methods. The method integrates pre-trained Arabic language models and architectural refinements, achieving improved retrieval and ranking accuracy for Arabic question answering. Why it matters: This work addresses the underrepresentation of Arabic in NLP research by providing a novel approach and publicly available code to improve Arabic text retrieval, which can benefit various applications like Arabic search engines and question-answering systems.
Researchers introduce a new task for generating question-passage pairs to aid in developing regulatory question-answering (QA) systems. The ObliQA dataset, comprising 27,869 questions from Abu Dhabi Global Markets (ADGM) financial regulations, is presented. A baseline Regulatory Information Retrieval and Answer Generation (RIRAG) system is designed and evaluated using the RePASs metric.
This paper introduces a new non-statistical Arabic lemmatizer algorithm designed for information retrieval systems. The lemmatizer leverages Arabic language knowledge resources to generate accurate lemma forms and relevant features. The algorithm achieves a maximum accuracy of 94.8% and 89.15% on first seen documents, outperforming the Stanford Arabic model's 76.7% on the same dataset. Why it matters: Accurate Arabic lemmatization is crucial for improving the performance of Arabic information retrieval systems, which can enhance access to Arabic language content.
Akhil Arora from EPFL presented a framework for AI-assisted knowledge navigation, focusing on understanding and enhancing human navigation on Wikipedia. The framework includes methods for modeling navigation patterns, identifying knowledge gaps, and assessing their causal impact. He also discussed applications beyond Wikipedia, such as multimodal knowledge navigation assistants and multilingual knowledge gap mitigation. Why it matters: This research has the potential to improve information systems by making online knowledge more accessible and navigable, especially for platforms like Wikipedia that serve as critical resources for global knowledge sharing.
MBZUAI Professor Chih-Jen Lin gave a keynote at the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval in Taipei. Lin's address, titled ‘On the “Rough Use” of Machine Learning Techniques’, focused on instances where machine learning techniques are employed inappropriately, using examples from graph representation learning and deep neural networks. He advocated for the development of high-quality, user-friendly software to improve the practical application of machine learning and mitigate misuse. Why it matters: Showcases MBZUAI's faculty expertise and contributions to the discussion on responsible AI research and deployment on a global stage.
Researchers proposed Utility-Aligned Embeddings (UAE), a new framework designed to enhance Retrieval-Augmented Generation (RAG) by merging the precision of LLM re-ranking with the efficiency of dense vector retrieval. UAE trains a bi-encoder to imitate an LLM utility distribution using a Utility-Modulated InfoNCE objective, injecting graded utility signals directly into the embedding space. On the QASPER benchmark, UAE improved retrieval Recall@1 by 30.59% and was over 180 times faster than efficient LLM re-ranking methods while preserving competitive performance. Why it matters: This approach offers a practical way to significantly improve the accuracy and speed of RAG systems by providing more reliable contexts at scale without heavy computational cost.